D&D 5th Edition: Thread #7

[PairO'Dice Lost]

Er, all of it? That's incredibly vague. Are you suggesting we use 3e or 4e? They're very different. And which two specific alterations? Removing 1/2 level bonus? Removing spells that bypass skills? Using "some method to remove the reliance on stat-boosters"?

I said that either 3e or 4e would work better; which one they'd prefer to start with doesn't really matter (I'm guessing 4e since it's not under OGL), but yes, all of it; strip out the spells, powers, feats, and other selectable fiddly bits and just look at the basics (chapters 1, 4, 8, 9, and 10 in the 3e PHB, and the equivalent in 4e). The math works--not in the marketing-speak "the 4e math just works!" sense, but in the sense that it doesn't collapse under scrutiny like bounded accuracy and the fact that the concept of the "sweet spot" exists at all. The things that make low and high levels not work in either edition are fragile PCs on the low end and spells, powers, and stacking bonuses on the high end, while the basic resolution mechanics work just fine.

Speaking of which, those were indeed the two suggestions: remove the item reliance and remove overly-high bonuses. If you're making a new edition, those two are as simple as not having items grant pluses and not having spells that grant +20 or more to certain rolls. You can do the former in several ways (give the PCs scaling bonuses to compensate for item plus removal, lower the monsters' stats, leave the numbers the same as in standard 3e or 4e and just make the game "harder" that way, etc.), which again depends on taste; any of them would work, so you can just throw darts at a dartboard if you can't decide.

Pas that, I'll just point to this article, which I'm sure many of you have already read, but it lists out the reasons for bounded accuracy better than I can.

I've already posted a rant on that article somewhere, so I'll just summarize to spare people reading it again:

Spoiler

Show

Getting better at something means actually getting better at something.
In 3e and 4e, you already get better at something when you level. If you're talking about objective benchmarks, pick some arbitrary benchmark (bashing in a door, hitting a troll, jumping a gap, whatever) and you get better at that as you level. If you're talking about relative skill, in 3e attack bonus outpaces AC by virtue of one of them scaling and the other not while in 4e it outpaces AC by virtue of bonus stacking. The fact that Mearls thinks that it isn't the case in either edition that "When a fighter gets a +1 increase to his or her attack bonus, it means he or she hits monsters across the board 5% more often" just means he can't do math; by definition, getting +1 on a d20 means your chances of hitting against X increase by 5% for any X as long as you need ≤20 to hit (as is the case with all published monsters).

Nonspecialized characters can more easily participate in many scenes. You don't need bounded accuracy to do this; as I mentioned above, simply instituting a +1/2 level rule in the 3e system makes the difference between an expert and a noob within the RNG if you get rid of the game-breaking spells.

The DM's monster roster expands, never contracts. This is already the case in 3e, as I think I mentioned upthread with the name level discussion, it's just that mid levels mean you can take on several hundred goblins instead of two dozen--which, I would argue, is a good thing. If facing whole armies is not to your liking, look to 4e, where a specific selling point of the edition was that minions let a party of 5 PCs take on 20 monsters and have it be a fair fight.

Bounded accuracy makes it easier to DM and easier to adjudicate improvised scenes. Page 31 in the 3.5 DMG, page 42 in the 4e DMG. Lists of benchmark DCs for improvisation have been around for a while.

It opens up new possibilities of encounter and adventure design. As mentioned, lots of mooks vs. few strong characters is already a thing.

It is easier for players and DMs to understand the relative strength and difficulty of things. For basic NPCs, there's this thing called a Monster Manual that Mearls may have heard of that stats out your basic hobgoblin. For "named" NPCs, I don't see why "plate armor = tough to hit, leather armor = easy to hit" should be a valid assumption when rogues and bards are running around.

It's good for verisimilitude. See above regarding skill benchmarks.

Suffice to say I find Mearls's arguments unconvincing at best and a sign that he hasn't read the previous two editions' DMGs at worst.

(Dice and Kurald's math seems to assume that a DM will call for a single Strength check to determine an arm wrestling contest, or a single Intelligence check to determine a chess match; this is probably sensible, since I've never heard of a group of D&D players who had the patience to play through a whole Skill Challenge-esque series of die rolls when they say "My character challenges the NPC Duke to a chess game."

Meanwhile, AgentPaper is referring to things like a grapple contest; while such checks certainly happen in D&D, it is important to realize that their individual success or failure is not usually important, because they are one roll in a series of many d20 rolls that resolve an entire battle.)

Note the difference here between a Strength check and a Strength contest. In both if your examples you're referring to opposed rolls, but the contest part isn't the problem; two fighters with max Str, max ranks/training, and identical miscellaneous bonuses rolling against each other will have a 50/50 shot regardless of the skill system being used. The check part is the problem, because that's the part that has the bad objective benchmarks and that's the part where comparing static DCs vs. the width of the RNG causes problems.

Flip it around and use normal skill checks: even if you need just one success on a DC 25 Int check in one case and multiple successes on multiple DC 25 Str checks in another case, if a 1st level character can succeed at a DC 25 check and accomplish "godlike" results, it doesn't matter how many checks they need to or can make because the fact that they can make it at all is the problem in and of itself.

Frankly, I think the way this needs to get resolved is to explicitly make separate mechanics for single-roll situations as opposed to many-roll situations. At the simplest level, what if Skill or Ability Checks with an important individual result were made using 1d10 or 3d6 instead of 1d20, while in-combat 1d20 rolls remained unchanged? Would that strike everyone as a relatively effective patch to the problem?

It's not an in-combat vs. out-of-combat divide, though. If you want a dragon to be noticeably better than a goblin, you need a wide enough gap for that to be the case whether it's the dragon attacking the goblin or vice versa, the dragon rolling Hide vs. the goblin's Spot or vice versa, or both of them rolling either against a third party. Again, it's the bonus spread relative to the width of the RNG, the bonus spread between opponents, and the DC benchmarks that matter, not the importance of individual rolls.

But why should the 20 str barbarian be winning against a 35 str purple worm? If the monster's stat bonus is that much higher than the Barbarian's, then that monster probably shouldn't be a level appropriate challenge for that barbarian. By the time the Barbarian is taking a Purple Worm on toe to toe, he should be as strong as the purple worm. The Worm will still have size modifiers on his side, but a dedicated grappling barbarian should have some ability that says "I ignore size modifiers"


What you describe isn't a problem with the 20str barbarian vs 5 str wizard, it's a problem with expectations that monsters are stronger than PCs.

As I mentioned in a later post, I agree with you completely and think that they should have normalized it so CR = HD and stats are in line for PCs and NPCs of similar HD and role. I was using the purple worm example to show that changing how the X Str barbarian interacts with the X-Y Str wizard will also change how the X Str purple worm interacts with the X-Y Str barbarian, even if it's something like wizard Str 16, barbarian Str 18, and purple worm Str 20, and thus you can't push for "the stronger party easily beats the weaker party" in one case and "the weaker party has a noticeable chance to escape" in the other.